J.C. Bradbury explains why he doesn’t like replacement level valuation. I find his last argument less than compelling:
Replacement-level terminology is unnecessarily complicated. If the sabermetric community wants to continue spreading influence beyond its small sphere of influence it can express all its important concepts in terms familiar to all baseball fans. When I played little league, we talked about on-base percentage and slugging percentage. It’s easy to explain DIPS through discussing strikeouts, walks, and homers (“see, the pitcher does these things all on his own, without fielding help”). But replacement level, why not just bring up quantum physics? This is baseball, it’s supposed to be fun. Replacement-level language is complicated and it adds no useful additional information to concepts that can be expressed more simply, especially in reference to league average. Replacement-level terminology should be rejected for parsimony alone.
I’m sorry, complicated is not an excuse. The runs created formula is complicated, but you can get someone to understand it in non-mathematical terms. Hell, the concept of an at bat is incredibly complicated, and can only be described by what it is not. (Go ahead, try explaining an at bat to a baseball novice.) I can explain the concept of replacement level without explaining the intricacies of the calculations, just like the runs created formula.
But you’re arguing JC’s point – concepts should be expressed without jargon and advanced mathematics.
The real problem is that when you explain the idea of replacement level talent, you’re explaining a flawed concept.
David –
I have to say I agree with Bradbury on this one. Most fans would have a problem with Replacement Level Player. On the other hand, they would easily understand League Average Player. I am not a sabermetrician. I am, however, very much into the results of sabermetric analysis. I much prefer league average comparisons to replacement level comparisons. Keep up the good work.
Replacement level is useful for evaluating performance in the context of monetary value.
If evaluating pure performance, I’d say that league average is a more useful and fan-friendly comparison.
I think that it ultimately comes down to whether or not you’re interested in the financial aspects of baseball.
Replacement level is useful for evaluating performance in the context of monetary value.
If evaluating pure performance, I’d say that league average is a more useful and fan-friendly comparison.
I think that it ultimately comes down to whether or not you’re interested in the financial aspects of baseball.
Additionally, what’s wrong with having two different reference points? They’re both solid, they’re both practical. I don’t imagine that the Sabermetric community is interested in adjusting their practice to achieve a wider following.
I don’t see why it is a big deal– if David Wright is out for the season his ~5 WAR gets replaced with likely 0 WAR. Easy.
If you’re looking at average, then his 3 wins above average get replaced by 2 wins below average.
In the former, the baseline makes sense, in the latter, you now have to explain where the -2 wins coems from.
It’s an easy concept. Maybe not as simple as average, but still easy. There are many freely available scrub-level players that teams can pay (approximately) league-minimum. So in order to to have any value to a team, you have to outperform that level of production.
— —
It’s also an *important* concept. Which is more valuable, -10 runs compared to average over 600 PAs or -5 runs over 300 PAs? -5 looks like a better number than -10, but think about who’s going to provide the extra 300 PAs for the second team. The best assumption is freely available talent, i.e. replacement level talent. That’s about -20 runs per 600 PAs or -10 runs per 300 PAs. So the 300 PA player becomes 1/2 of a -15 run per 600 PA position on the team.
In other words, the value of a player is relative to some baseline level of production. You can’t assume that baseline level to be any better than replacement level, because better production costs money.